This five-year study proposes to develop and evaluate an educational intervention to prevent STDs among sexually active inner-city youth using the services of two Boston adolescent clinics. (Boston Medical Center and Children's Hospital). A randomized controlled trial involving an experimental group (N=1938) and a routine care group (N=1938) will be used to test the efficacy of a 15-minute interactive video in combination with a 15-minute interactive counseling session delivered by a trained health educator. Several characteristics distinguish this study. First, the intervention will be designed to deliver individualized content depending upon the adolescent's age, gender, intentions with respect of sexual activity (secondary abstinence or continued sexual activity), and stage of change with regard to adoption of consistent condom use. The individualization takes place within both the video and counseling components of the intervention. Second, evaluation of the intervention will be based on a biological marker for consistent condom use- Chlamydia infection-along with behavioral measures, as opposed to self- reported behavioral measures alone. The intervention builds upon Social Cognitive Theory, and person-to-person counseling. The video component will involve multiple modules tailored to the characteristics of participants. Within the interactive video, the participant will be free to select the sequence of content according to the salience of topics. The structure of the counseling session reflects this approach as well. The participant selects from a menu of topic and determines the order with which content is discussed. Third, the intervention will use booster sessions to enhance effectiveness over time. Fourth, intervention effectiveness will be assessed at six-week, six-month, and one-year follow-up visits. Non-invasive (urine) methods will be used to determine disease (Chlamydia) status at each follow-up. RNA messenger analysis will be used at six week follow-up to distinguish between active Chlamydia infection and residual DNA, for those participants with infection at enrollment. Fifth, a computer-based data collection system will be developed to enhance the validity of self-reported behavioral and attitudinal measures. Our hypothesis is that, relative to controls, intervention participants will evidence at 6-month and one-year follow-up a 25% differences (12% versus 9%) in Chlamydia infection rates.